© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 4450
    Prev
    Next
    INDEX
    Explanations

    mentions of specific products and their characteristics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    481
    +0.31
    1.5%
    87
    +0.13
    0.6%
    620
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.31
    0.08
    153
    +0.13
    0.04
    45
    +0.09
    0.02
    Negative Logits
    illance
    -0.82
    protection
    -0.67
     confisc
    -0.64
     eaves
    -0.63
    pressure
    -0.63
    ÄŁ
    -0.63
     blackmail
    -0.63
    prison
    -0.63
    obil
    -0.62
     Telephone
    -0.60
    POSITIVE LOGITS
     contenders
    1.18
     ones
    1.14
     favorites
    1.08
     outl
    1.00
     notable
    0.99
     standout
    0.99
     finalists
    0.98
     categories
    0.97
     classics
    0.96
     noteworthy
    0.92
    Activations Density 0.842%

    No Known Activations